On https://syzkaller.appspot.com/bug?extid=4c0d0c4cde787116d465

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hillf,

first of all I should say that I'm new to bluetooth and may misunderstand something.

IIUC your patch at https://syzkaller.appspot.com/text?tag=Patch&x=15faf610e80000
assumes that an instances of 'struct sco_conn' can share the same 'struct sock'
(that's why an extra calls to 'sock_hold()' was added). OTOH my patch at
https://lore.kernel.org/linux-bluetooth/20240403142706.25748-1-dmantipov@xxxxxxxxx/T/#t
assumes that this is wrong because SCO (by definition) is a point-to-point
link between the master device and a specific slave device, and prevents
from creating such a sharing instead. So the question is: should we always
assume 1:1 relationship between SCO connection and kernel socket?

Any comments are highly appreciated, thanks in advance.

Dmitry




[Index of Archives]     [Bluez Devel]     [Linux Wireless Networking]     [Linux Wireless Personal Area Networking]     [Linux ATH6KL]     [Linux USB Devel]     [Linux Media Drivers]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [Big List of Linux Books]

  Powered by Linux